Downloaded from http://bjo.bmj.com/ on November 26, 2014 - Published by group.bmj.

com

Clinical science

Collagen crosslinking in the management of

advanced non-resolving microbial keratitis
Rohit Shetty, Harsha Nagaraja, Chaitra Jayadev, Yathish Shivanna, Thungappa Kugar
Narayana Nethralaya Eye
Hospital, Bangalore,
Karnataka, India
Correspondence to
Dr Harsha Nagaraja, Narayana
Nethralaya Eye Hospital, 121/
C, Chord Road, 1 R Block,
Rajajinagar, Bangalore 560
010, Karnataka, India;
harshdr@gmail.com
Received 16 January 2014
Revised 24 February 2014
Accepted 16 March 2014
Published Online First
7 April 2014

ABSTRACT
Aim To evaluate the efcacy and safety of corneal
collagen crosslinking (CXL) in the management of culture
proven microbial keratitis.
Methods 15 eyes of 15 patients of microbial keratitis
were included in the study. Nine patients had bacterial
keratitis and six had fungal keratitis. All patients
underwent microbiological evaluation to identify the
causative organism. The depth of the inltrate was
determined clinically with slit lamp and measured
manually using anterior segment optical coherence
tomography. Patients were treated with antibiotics/
antifungals and those who did not respond to at least
2 weeks of topical medications underwent CXL as per
the standard protocol. The same preoperative topical
medications were continued post-CXL. All patients were
followed up every third day and observed for signs of
resolution of microbial keratitis.
Results Six of nine patients with bacterial keratitis and
three of six patients with fungal keratitis resolved
following CXL treatment. Patients with deep stromal
keratitis or endothelial plaque failed to resolve. All
patients had resolution of pain on the rst postoperative
day. There was an appearance of or increase in
hypopyon in seven patients. No intraoperative or
postoperative complications were noticed.
Conclusions CXL appears to be an effective procedure
in the management of supercial microbial keratitis. It
can be used as an adjunctive treatment in the
management of non-resolving microbial keratitis.

INTRODUCTION

To cite: Shetty R,
Nagaraja H, Jayadev C,
et al. Br J Ophthalmol
2014;98:10331035.

Corneal diseases are a major cause of vision loss

and blindness according to WHO; ocular trauma
and corneal ulceration result in 1.52 million new
cases of corneal blindness annually.1 Microbial keratitis is characterised by a corneal epithelial defect
with underlying stromal inammation caused by
replicating microorganisms. Microbial keratitis is a
leading cause of ocular morbidity and blindness
worldwide.2 Microbial keratitis requires aggressive
management to halt the disease process and reduce
the extent of corneal scarring, which leads to loss
of vision.3 4
It has been estimated that about 50% of the eyes
have poor visual outcome if the diagnosis and
initiation of appropriate antimicrobial treatment
are delayed.5 Long-term and intensive topical
chemotherapy has led to the emergence of
multidrug-resistant bacteria that might further complicate treatment. Microbes develop resistance to
antibiotics as a result of chromosomal mutation,
inductive expression of latent chromosomal genes,
exchange of genetic material via transformation,
bacteriophage transduction or plasmid conjugation.6

The high cost of drugs, high frequency of resistance

to antibiotics, and risk of corneal melting and
corneal scars make the choice of new approaches
very desirable. Since 2000, riboavin has been used
as a photosensitiser to inactivate pathogens in
plasma, platelet and red cell products.7 8 In vitro
experiments have proven the bactericidal effect
using a 365 nm ultraviolet-A (UV-A) light photo
activation of riboavin.9 10 Multiple studies have
indicated that this could be a new tool in the management of infectious keratitis resistant to antibiotic
treatment.11 In our study, we evaluated the role of
corneal collagen crosslinking (CXL) in the management of advanced non-resolving microbial keratitis.

MATERIALS AND METHODS

Patients with non-resolving microbial keratitis were
included in the study after taking informed
consent. The study was performed under the
Tenets of the Declaration of Helsinki. At initial
presentation, all patients underwent corneal scrapping for Grams stain, potassium hydroxide preparation (KOH wet mount), bacterial culture (blood
agar and chocolate agar) and fungal culture
(Sabouraud dextrose agar). Patients were started on
topical antibiotics/topical antifungals based on the
Grams stain/KOH wet mount reports which were
obtained immediately after the patient presented to
us. There was no delay in starting topical medications in any of the patients. All patients were prescribed topical atropine 1% eye ointment in
addition to antibacterials and antifungals. They
were followed up every third day and observed for
signs of resolution of the corneal involvement and
hypopyon. Those patients not responding to
topical chemotherapy for more than 2 weeks were
considered as having non-resolving microbial keratitis involvement and advised to undergo CXL.
Patients with culture proven bacterial and fungal
keratitis were included in the study. Those patients
with perforated corneal ulcer, endophthalmitis,
viral keratitis and pregnancy were excluded.
Procedure: Topical anaesthesia was administered
using 0.5% proparacaine drops. Epithelium was not
removed as there was already a large epithelial defect
overlying the area of keratitis in all patients.
Riboavin drops (Medio-Cross riboavindextran
solution, 0.1%) were instilled topically on the cornea
every 2 min for a period of 30 min. The cornea was
then illuminated using a Phoenix UV-A system
(Peschke
Meditrade
GmbH,
Huenenberg,
Switzerland), UVA 365 nm, with an irradiance of
3 mW/cm2 and a total dose of 5.4 J/cm2 over 30 min.
During the period of UV illumination, riboavin was
administered to the eye every 2 min. Patients were
continued with the same topical medications that

Shetty R, et al. Br J Ophthalmol 2014;98:10331035. doi:10.1136/bjophthalmol-2014-304944

1033

Downloaded from http://bjo.bmj.com/ on November 26, 2014 - Published by group.bmj.com

Clinical science
were being used prior to CXL. Topical corticosteroids or NSAIDs
were not prescribed after the crosslinking procedure.

RESULTS
In all, 15 eyes of 15 patients including 10 men and ve women
with culture proven non-healing microbial keratitis were
included in the study. Mean age of the patients was 51
13.38 years (range 2771 years). Based on the smear and
culture reports, nine patients had bacterial keratitis and six had
fungal keratitis. The most common bacteria isolated were
staphylococci and aspergillus was the most common fungus.
Seven patients had hypopyon at the time of presentation.
Table 1 gives the pre-CXL and post-CXL details of all patients.
All patients who underwent CXL had resolution of pain on
the rst postoperative day. In four patients with hypopyon
cornea ulcer, there was an increase in hypopyon by the third
postoperative day. There was an appearance of hypopyon in
three patients who did not have hypopyon prior to CXL
(gures 14). In patients who responded to CXL (nine out of
15), the mean time for epithelial healing was 21.36.14 days
and the mean time for resolution of corneal inltrate was 33.44
6.2 days. Of the seven patients who either had an increase or
appearance of hypopyon, three patients underwent therapeutic
keratoplasty while in the other four, the resolution of hypopyon
took longer than the resolution of the corneal inltrate.
The mean preoperative score on the Wong-Baker FACES Pain
Rating Scale was 8.270.73 and the mean postoperative score was
0.530.85 indicating that there was a signicant reduction in pain.
No intraoperative complications were noted.

DISCUSSION
In patients undergoing therapeutic keratoplasty for non-healing
microbial keratitis, approximately a third of transplanted grafts

Figure 1 Preoperative non-resolving bacterial keratitis with anterior

stromal inltrate and hypopyon.
become re-infected.12 There has therefore been a constant quest
to nd an alternative to therapeutic keratoplasty when medical
management fails.
Tsugita et al13 showed that the combination of riboavin and
UVA caused a deactivation of the RNA in tobacco mosaic virus.
Martins et al opined that a combination of UVAriboavin has
antibacterial properties in vitro against microorganisms causing
microbial keratitis.9 Hence, we evaluated the effectiveness of
CXL in our patients with infective keratitis. We found that in
patients with supercial corneal inltrates involving only the
anterior third or less of the stroma, there was a better response
than in those cases with deeper inltrates.
According to previous studies, the treatment seems to be most
effective in blocking corneal melt caused by Gram-negative bacteria (92%) followed by Gram-positive bacteria (84%), acanthamoeba (71%) and nally fungi (61%).1420 In our experience,
bacterial keratitis has a better prognosis compared with fungal

Table 1 Details of patients with non-resolving microbial keratitis showing both precollagen and postcollagen crosslinking parameters
Age/
gender

Clinical
diagnosis

Microbiology
reports

Depth of
infiltrate

36/M
42/M

G +ve
G ve

Staph.
Pseudomonas

68/F
52/M
47/M
65/M

G
G
G
G

Staph.
Strepto.
Staph.
Pseudomonas

Superficial stroma
Full thickness ring
infiltrate
Surface infiltrate
Full thickness
Superficial stromal
Anterior stroma

71/F

G +ve

Staph.

36/M

G ve

Inconclusive

53/M
27/F

G +ve
Fungal

Staph.
Aspergillus

58/M

Fungal

Candida

65/F

Fungal

Fusarium

43/M

Fungal

54/F
48/M

Fungal
Fungal

+ve
+ve
+ve
ve

Anterior third of
stroma
Full thickness

Hypopyon

Post-CXL response

No
5 mm

Healed with scar

Non-resolving;
increase in hypopyon
Healed with scar
Non-resolving
Healed with scar
Healed with scar;
hypopyon
Healed with scar

No
4 mm
No
No
No
3 mm

Superficial stromal
Anterior third of
stroma
Deep stromal and
endothelial plaque
Superficial +
mid-deep stroma

No
5 mm

Aspergillus

Anterior stroma

1 mm

Candida
Aspergillus

Deep stroma
Anterior third
stroma

2 mm
No

2 mm
No

Non-resolving;
increase in hypopyon
Healed with scar
Healed with scar
Increase in hypopyon
Non-resolving
Partial scar;
non-resolving;
hypopyon
Healed with scar;
increase in hypopyon
Non-resolving
Healed with scar;
hypopyon

Time for
epithelial
healing

Time for resolution

of infiltrate

Time for
resolution of
hypopyon

14 days

4 weeks

No hypopyon
Non-resolving

18 days

15 days
28 days

4 weeks

4 weeks
5 weeks

No hypopyon
Non-resolving
No hypopyon
7 weeks

21 days

5 weeks

Non-resolving

14 days
26 days

4 weeks
6 weeks

No hypopyon
8 weeks

Non-resolving

Non-resolving

28 days

Only the superficial

infiltrate resolved in
5 weeks
6 weeks

8 weeks

28 days

5 weeks

Non-resolving
6 weeks

G +ve- Gram positive; G -ve- Gram negative; Staph.- Staphylococci species; Strepto.- Streptococci species.

1034

Shetty R, et al. Br J Ophthalmol 2014;98:10331035. doi:10.1136/bjophthalmol-2014-304944

Downloaded from http://bjo.bmj.com/ on November 26, 2014 - Published by group.bmj.com

Clinical science

Figure 4 Post-CXL at 7 weeks showing complete resolution of

corneal inltrate with appearance of hypopyon.
Figure 2 Post-CXL at 8 weeks showing complete resolution of
corneal inltrate with persisting hypopyon.
keratitis probably because of the fact that most of the bacterial
keratitis involved only the supercial part of the corneal stroma.
We also noticed that all patients who underwent CXL had
resolution of pain on the rst postoperative day. The possible
explanation for this could be damage to the subepithelial nerve
plexus by the riboavinUVA combination due to chemical
denervation. The reduction of corneal sensation in all patients
post-CXL supports this hypothesis.
The other important nding was that there was an increase in
hypopyon in four patients post-CXL and appearance of hypopyon in three patients. We attribute this increase or appearance of
hypopyon possibly due to the penetration of riboavin through
the inamed and oedematous stroma into an already inamed
anterior chamber causing further inammation. It was also
noted that the hypopyon took a longer time to resolve than the
re-epithelisation and resolution of surface inltrate.
The main drawback of the study was that it was not known if
the cases of microbial keratitis would have responded to topical
medications if they were continued for a longer period of time;
however, CXL did accelerate healing once it was performed.
In conclusion, CXL appears to be an effective procedure in
treating non-resolving microbial keratitis with supercial stromal
involvement. CXL can be an effective adjunctive treatment and
add to the armamentarium of treatment modalities in the management of resistant microbial keratitis. However, further
studies are required to streamline the indications, protocol and
safety prole of CXL in the management of microbial keratitis.

Contributors All authors have contributed in the preparation of the manuscript

and the clinical study.
Competing interests None.
Ethics approval Narayana Nethralaya Ethics Committee.
Provenance and peer review Not commissioned; externally peer reviewed.

REFERENCES
1
2

3
4
5
6
7
8
9

10

11

12

13
14
15
16
17
18

19

Figure 3 Non-resolving fungal keratitis after 2 weeks of treatment

with topical antifungals.

20

Shetty R, et al. Br J Ophthalmol 2014;98:10331035. doi:10.1136/bjophthalmol-2014-304944

Whitcher JP, Srinivasan M, Upadhyay MP. Corneal blindness: a global perspective.

Bull World Health Organ 2001;79:21421.
Burd EM, Ogawa GSH, Hyndiuk RA. Bacterial keratitis and conjunctivitis. In:
Smolin G, Thoft RA, eds. The cornea scientic foundations and clinical practice. 3rd
edn. Boston: Little, Brown, & Co, 1994:11567.
Keay L, Edwards K, Naduvilath T, et al. Microbial keratitis predisposing factors and
morbidity. Ophthalmology 2006;113:10916.
Wong T, Ormonde S, Gamble G, et al. Severe infective keratitis leading to hospital
admission in New Zealand. Br J Ophthalmol 2003;87:11038.
Jones DB. Decision-making in the management of microbial keratitis.
Ophthalmology 1981;88:81420.
BertinoJ S. Impact of antibiotic resistance in the management of ocular infections:
the role of current and future antibiotics. ClinOphthalmol 2009;3:50721.
Goodrich RP. The use of riboavin for inactivation of pathogens in blood products.
Vox Sang 2000;78:21115.
McAteer MJ, Tay-Goodrich B, Doane S. Photoinactivation of virus in packed red
blood cell units using riboavin and visible light. Transfusion 2000;3(Suppl):99S.
Martins SA, Combs JC, Noguera G, et al. Antimicrobial efcacy of riboavin/UVA
combination (365 nm) in vitro for bacterial and fungal isolates: a potential new
treatment for infectious keratitis. Invest Ophthalmol Vis Sci 2008;49:34028.
Schrier A, Greebel G, Attia H, et al. In vitro antimicrobial efcacy of riboavin and
ultraviolet light on Staphylococcus aureus, methicillin-resistant Staphylococcus
aureus, and Pseudomonas aeruginosa. J Refract Surg 2009;25:S799802.
Makdoumi K, Backman A, Mortensen J, et al. Evaluation of antibacterial efcacy of
photo-activated riboavin using ultraviolet light (UVA). Graefes Arch Clin Exp
Ophthalmol 2010;248:20712.
Shi W, Liu M, Gao H, et al. Perioperative treatment and prognostic factors for
penetrating keratoplasty in Acanthamoeba keratitis unresponsive to medical
treatment. Graefes Arch Clin Exp Ophthalmol 2009;247:13838.
Tsugita A, Okada Y, Uchara K. Photosensitized inactivation of ribonucleic acids in
the presence of riboavin. Biochim Biophys Acta 1965;103:3603.
Anwar HM, El-Danasoury AM, Hashem AN. Corneal collagen crosslinking in the
treatment of infectious keratitis. Clin Ophthalmol 2011;5:127780.
Makdoumi K, Mortensen J, Crafoord S. Infectious keratitis treated with corneal
crosslinking. Cornea 2010;29:13538.
Morn H, Malmsj M, Mortensen J, et al. Riboavin and ultraviolet a collagen
crosslinking of the cornea for the treatment of keratitis. Cornea 2010;29:1024.
Iseli HP, Thiel MA, Hafezi F, et al. Ultraviolet A/riboavin corneal cross-linking
for infectious keratitis associated with corneal melts. Cornea 2008;27:5904.
Khan YA, Kashiwabuchi RT, Martins SA, et al. Riboavin and ultraviolet light a
therapy as an adjuvant treatment for medically refractive Acanthamoeba keratitis:
report of 3 cases. Ophthalmology 2011;118:32431.
Martins SA, Combs JC, Noguera G, et al. Antimicrobial efcacy of riboavin/UVA
combination (365 nm) in vitro for bacterial and fungal isolates: a potential new
treatment for infectious keratitis. Invest Ophthalmol Vis Sci 2008;49:34028.
Price MO, Tenkman LR, Schrier A, et al. Photoactivated riboavin treatment of infectious
keratitis using collagen cross-linking technology. J Refract Surg 2012;28:70613.

1035

Downloaded from http://bjo.bmj.com/ on November 26, 2014 - Published by group.bmj.com

Collagen crosslinking in the management of

advanced non-resolving microbial keratitis
Rohit Shetty, Harsha Nagaraja, Chaitra Jayadev, Yathish Shivanna and
Thungappa Kugar
Br J Ophthalmol 2014 98: 1033-1035 originally published online April 7,
2014

doi: 10.1136/bjophthalmol-2014-304944
Updated information and services can be found at:
http://bjo.bmj.com/content/98/8/1033

These include:

References
Email alerting
service

Topic
Collections

This article cites 18 articles, 3 of which you can access for free at:
http://bjo.bmj.com/content/98/8/1033#BIBL
Receive free email alerts when new articles cite this article. Sign up in the
box at the top right corner of the online article.

Articles on similar topics can be found in the following collections

Cornea (464)
Ocular surface (553)
Eye (globe) (632)
Neurology (1220)

Notes

To request permissions go to:

http://group.bmj.com/group/rights-licensing/permissions
To order reprints go to:
http://journals.bmj.com/cgi/reprintform
To subscribe to BMJ go to:
http://group.bmj.com/subscribe/